I am working on a project where I am trying to find the KDE for a bunch of points. Currently I am using kde.R to find KDE using the standard normal kernel. I want to use different kernels and compare the results. However, while I have found a lot of libraries for univariate data that allows me to pick a certain kernel, this doesn't seem to exist for bivariate data.
Does anyone know if such a function/library exist?
If not, I'm only left with the option of writing my own kernel code!